RightNow-AI/local-kimi

Optimized local serving engine for Kimi-Linear-48B: INT4 quantizer, fused decode kernels for a measured 3.18x, and an OpenAI-compatible server. Ships with k3, a bridge that detects the client per request so Claude Code, Codex, Cline, Aider and opencode all work unchanged.

4 best points · 0 comments · 1 Hacker News post · 44 GitHub stars · View on GitHub

Hacker News discussions

• Kimi k3 run on RTX 5090 (4 points, 0 comments, by OsamaJaber, 2026-07-30)